A Daytime Measurement of the Lunar Contribution to the Night Sky Brightness in LSST's ugrizy Bands-- Initial Results

TL;DR

This study measures the lunar contribution to night sky brightness at the LSST site using daytime sunlight proxies, providing insights into lunar light scattering and aiding astronomical observation planning.

Contribution

First measurement of lunar night sky brightness contribution across LSST bands using daytime proxies and correction for spectral differences.

Findings

Lunar contribution to night sky brightness varies with zenith and lunar angle.

Estimated lunar brightness uncertainty is about 10%.

Results align with previous estimates of scattered lunar photons.

Abstract

We report measurements from which we determine the spatial structure of the lunar contribution to night sky brightness, taken at the LSST site on Cerro Pachon in Chile. We use an array of six photodiodes with filters that approximate the Large Synoptic Survey Telescope's {\it u, g, r, i, z,} and {\it y} bands. We use the sun as a proxy for the moon, and measure sky brightness as a function of zenith angle of the point on sky, zenith angle of the sun, and angular distance between the sun and the point on sky. We make a correction for the difference between the illumination spectrum of the sun and the moon. Since scattered sunlight totally dominates the daytime sky brightness, this technique allows us to cleanly determine the contribution to the (cloudless) night sky from backscattered moonlight, without contamination from other sources of night sky brightness. We estimate our uncertainty in the relative lunar night sky brightness vs. zenith and lunar angle to be 10\,\%. This information is useful in planning the optimal execution of the LSST survey, and perhaps for other astronomical observations as well. Although our primary objective is to map out the angular structure and spectrum of the scattered light from the atmosphere and particulates, we also make an estimate of the expected number of scattered lunar photons per pixel per second in LSST, and find values that are in overall agreement with previous estimates.